Approximation for the minimal test cover / minimal group test problem

25 Views Asked by At

There are multiple approximation methods I find for the minimal test cover, where approximation is with respect to the size of the test set. However I am looking for approximation which starts with a set size, and finds a set of tests which as the best possible in terms of percent of objects resolved. Given: Set of Objects S = s1 ... sN, set of binary tests t1 .. tK and a desired size of minimal-test-set M. Find such set of M tests out of much larger collection of K tests which allows to identify the maximal fraction of the original set of objects S, leaving the rest in un-resolved groups.